Brain-derived neurotrophic factor modulates hippocampal synaptic transmission by increasing N-methyl-D-aspartic acid receptor activity

Abstract

Neurotrophins (NTs) have recently been found to regulate synaptic transmission in the hippocampus. Whole-cell and single-channel recordings from cultured hippocampal neurons revealed a mechanism responsible for enhanced synaptic strength. Specifically, brain-derived neurotrophic factor augmented glutamate-evoked, but not acetylcholine-evoked, currents 3-fold and increased N-methyl-D-aspartic acid (NMDA) receptor open probability. Activation of trkB NT receptors was critical, as glutamate currents were not affected by nerve growth factor or NT-3, and increased open probability was prevented by the tyrosine kinase inhibitor K-252a. In addition, the NMDA receptor antagonist MK-801 blocked brain-derived neurotrophic factor enhancement of synaptic transmission, further suggesting that NTs modulate synaptic efficacy via changes in NMDA receptor function.

Figure 1

BDNF enhances responsiveness to iontophoretically-applied glutamate or NMDA but not acetylcholine or AMPA. (A) Example traces of control response to glutamate iontophoresis (10-ms pulse; indicated by arrowhead) and response after 20 min of BDNF exposure. Holding potential was −40 mV. The iontophoresis pipette was located in the dendritic region of pyramidal cells. Traces in this and subsequent figures represent the average of five consecutive sweeps. (B) Time course of mean effect of BDNF on glutamate (■) or acetylcholine (▴) responsiveness (mean ± SEM). (C) Example traces of control response to NMDA iontophoresis and response after 20 min of BDNF exposure. (D) Example traces of control response to AMPA iontophoresis and response after 20 min of BDNF exposure. (E) Group data for effects of BDNF on responses to glutamate (P < 0.01; n = 7), NMDA (P < 0.05; n = 7 cells), AMPA (P > 0.4; n = 5), or acetylcholine (P > 0.4; n = 4).

Figure 2

BDNF increases NMDA receptor P_o. (A) Example sweeps from a cell-attached single-channel recording before and during application of BDNF (20 ng/ml). Sweeps came from time points indicated by arrows in B. This patch contained two channels. The holding potential was −80 mV; solutions are described in Methods. (B)Time course for effect of BDNF on channel P_o. (C) Acute exposure to BDNF significantly enhanced NMDA receptor channel P_o (P < 0.01; n = 9), normalized according to the number of channels in the patch. K-252a (200 nM) completely blocked the effect of BDNF (P > 0.3; n = 6). (D) Current-to-voltage relationships were not different between BDNF-treated cells (n = 12) and control cells (n = 9). Treated cells were preexposed to BDNF for 15 min. (E) The mean number of openings per sec was increased in BDNF-treated cells (P < 0.05). (F) Distribution of open times from summed data of all control cells (n = 8 recordings). Values are plotted on square-root/log coordinates and fitted with the maximum likelihood method (24). Dashed lines represent individual components (1.47 ms and 4.91 ms), solid line represents sum. In these recordings, a briefer time component may be missed due to the relatively long sampling interval used to maximize the total observation time per sweep. (G) Summed distribution from all BDNF-treated cells (n = 10 recordings). Dashed lines represent individual components (1.71 ms and 3.51 ms), solid line represents sum. The shapes of the control and BDNF-treated distributions were not significantly different (P > 0.1, Kolmogorov–Smirnov test), nor were mean open times averaged from individual experiments (control: 1.69 ± 0.28 and 4.69 ± 0.39 ms, BDNF-treated: 1.71 ± 0.29 and 3.98 ± 0.41 ms; P > 0.2).

Figure 3

Glutamate responsiveness is enhanced by BDNF, but not NT-3 or NGF. BDNF-induced increase in synaptic activity is blocked by NMDA receptor antagonism. (A) Example traces of control response to glutamate iontophoresis (10 ms pulse) and response after 20 min of NT-3 exposure. (B) Time course of mean effect of NT-3 on glutamate responses. (C) Group data for effects of heat-inactivated BDNF (HI-BDNF; P > 0.1; n = 3), BDNF (P < 0.01; n = 7), NGF (P > 0.4; n = 10), and NT-3 (P > 0.1; n = 11) on glutamate responsiveness. (D) Example traces of baseline synaptic activity and activity after 20 min of BDNF exposure, in the presence or absence of the NMDA receptor antagonist MK-801 (10 μM). (E) Group data for effect of BDNF on synaptic charge in the absence (P < 0.01; n = 11) or presence (P > 0.3; n = 7) of MK-801.